Table 4 Per-class precision, recall, F1-score, and AUC for multi-class classification.
From: Trade-offs between machine learning and deep learning for mental illness detection on social media
Class | Precision | Recall | F1-Score | AUC |
|---|---|---|---|---|
Anxiety (0) | 0.7947 | 0.8358 | 0.8147 | 0.98 |
Bipolar (1) | 0.8809 | 0.8140 | 0.8462 | 0.98 |
Depression (2) | 0.7158 | 0.7016 | 0.7086 | 0.90 |
Normal (3) | 0.9078 | 0.9202 | 0.9140 | 0.99 |
Personality Disorder (4) | 0.8750 | 0.5612 | 0.6838 | 0.97 |
Stress (5) | 0.6331 | 0.5989 | 0.6155 | 0.96 |
Suicidal (6) | 0.6613 | 0.7011 | 0.6806 | 0.92 |
